A stochastic segment model for phoneme-based continuous speech recognition

Improvements in the stochastic segment model for Phoneme recognition

“…Although hidden Markov models (HMMs) are currently one of the most successful approaches to acoustic modelling for continuous speech recognition, their performance is limited in part became of the assumption that observation features at different times are conditionally independent given the underlying state sequence and because the Markov assumption on the state sequence may not adequately model time structure. An alternative model, the stochastic segment model (SSlVl), was proposed to overcome some of these deficiencies [Roucos and Dunham 1987, Ostendorf and Roucos 1989, Roucos et al 1988.…”

Fast search algorithms for connected phone recognition using the stochastic segment model

“…The Stochastic Segment Model [6,10,9] was proposed as an alternative to hidden Markov models (HMMs), in order to overcome the limiting assumptions of the latter that observation features are conditionally independent given the underlying state sequence. The main disadvantage of the SSM over other methods is its computational complexity, which can be attributed to the fact that dropping the conditional independence assumption on the state sequence increases the size of the effective state space.…”

Formalizing Knowledge Used in Spectrogram Reading: Acoustic and Perceptual Evidence from Stops

“…Acoustic events are either present or absent, often extend over both time and frequency, and may occur simultaneously. Researchers have argued that * such acoustic events are hard to capture in systems which perform a frame-by-frame time analysis of the speech signal (Roucos and Dunham, 1987). In order to have the computer mimic the reasoning of spectrogram readers, one needs a system that can deal with qualitative measures in a meaningful way.…”